Research Structure Background

"Prevention and Control of Avian Influenza in the U.S." Research Structure

BACKGROUND AND SIGNIFICANCE
Influenza viruses. Influenza A viruses are divided into subtypes based on the antigenic properties of their hemagglutinin (HA) and neuraminidase (NA) surface glycoproteins (54). To date, 15 HA and 9 NA subtypes have been identified (115). The segmented nature of the virus genome, which consists of eight different molecules of single stranded RNA, allows reassortment of genes when a susceptible host is co-infected with two different influenza virus subtypes. Thus, influenza viruses with many combinations of HA and NA subtypes are routinely isolated from birds (79).

Interspecies transmission and virulence determinants in influenza virus . Wild aquatic birds (Orders Anseriformes and Charadriiformes) are considered the primary hosts of influenza A, in which the virus is enzootic (115). In these birds, influenza viruses usually replicate in the intestinal tract, cause no disease, and spread by fecal contamination of the water habitat (89). These viruses occasionally infect land-based birds (Order Galliformes) causing epizootics. Numerous examples of transmission of influenza from ducks to other species have been recorded. However, these events remain relatively rare, considering the high frequency of exposure of land-based birds as a result of co-mingling with waterfowl shedding influenza. It is thought that successful replication and spread of duck influenza in other species is restricted at multiple levels (23). The role of the HA protein in host range has been studied in some detail (28-30, 62, 63, 76, 106). The HA protein binds to sialic acid receptors on the cell surface. There are differences on the receptor specificity of influenza viruses between wild birds, land-based poultry and mammals. Depending upon how sialic acid receptors are presented on the host cell surface and how the HA protein interacts with them, an influenza virus may or may not start a successful cycle of infection (62). The recognition of sialic acid by HA is one of the critical determinants of host range.

In land-based poultry, influenza A infections range from asymptomatic to lethal. Virulence for chickens and other poultry can only result from infections with influenza A viruses that replicate efficiently in these species; i.e. that have crossed the species barrier. Viral determinants of virulence consist of individual viral components interacting with one another and with elements of the host cell, the immune system of the host, and the environment (2, 74, 122). The importance of efficient replication and spread of influenza A virus in a new host, such as land-based poultry, cannot be overemphasized; it is a pre-requisite for the expression of virulence. Multiple genes that determine replication efficiency are critical determinants of host range.

The cleavability of the HA protein plays a critical role in the pathogenicity of avian influenza viruses because it restricts tissue tropism (43, 44, 47, 49, 77, 94, 107). HA is synthesized as a precursor, which requires post-translational cleavage by host proteases to create functional HA and produce infectious virus particles (87). All highly pathogenic avian influenza viruses (HPAI) identified to date differ from their low pathogenic (LPAI) counterparts in the susceptibility of the HA to host proteases. HPAI are characterized by HAs that are highly susceptible to cleavage by numerous host proteases. In contrast, LPAI HA requires specific active proteases -such as trypsin- for cleavage and activation of infectivity.

Theoretically, viruses of all HA subtypes have the potential to cause disease. The extremely virulent viruses cause highly pathogenic avian influenza (HPAI), in which mortality may be as high as 100%. These viruses have been restricted to subtypes H5 and H7, although not all viruses of these subtypes are necessarily HPAI (103). Viruses with any other subtype cause a mild, primarily respiratory disease in poultry, which may be exacerbated by other infections or environmental conditions. Unlike other HA subtypes, the highly virulent H5 and H7 viruses possess multiple basic amino acids at the cleavage site of the HA. Since 1959, 21 primary outbreaks of HPAI in poultry have been reported (12 since 1990). Influenza viruses have been shown to infect all types of domestic or captive birds throughout the world. The probability of primary infections is proportional to the frequency of contact with feral birds. Provided that the virus can be transmitted horizontally in poultry, secondary virus spread is usually associated with human involvement, probably by transferring infective feces to susceptible birds (3). Currently there are suggestions that all highly pathogenic viruses are derived from low-pathogenic H5 and H7 viruses. The exact molecular mechanisms involved in this transformation are poorly understood, although it is known that the acquisition of multiple basic amino acids at the cleavage site of the HA protein play a major role. Hosts that favor this transformation and contributing environmental factors are poorly characterized, although chickens appear to provide an ideal environment.

We know much more about the molecular features of HA and NA that are associated with virulence than about the genome-wide process of molecular adaptation that enable an aquatic avian influenza virus to transmit from ducks to other species. Although adaptive changes in the surface and internal genes of waterfowl viruses are thought to be required for interspecies transmission, they remain to be defined. Phylogenetic analysis of NP gene segments has established that there are five host-specific lineages and suggests the putative role of NP as one determinant of host range (33). The other internal genes (PB1, PB2, PA, M1, M2, NS1, and NS2) may also play similar roles. The involvement of polymerase gene products in host range and pathogenicity has been demonstrated in a number of studies (33, 38, 56, 80, 91, 116). NS gene products have been implicated in host range as well (15, 22, 92). The viral NS1 protein has multiple functions that counteract cellular antiviral activity (31, 53, 67). NS1 inhibits NF-_B and IRF3 transcription factors in mammalian cells, both of which are critical for the synthesis of type I interferons (IFN-I) (11, 32, 105, 110). In addition, NS1 inhibits the dsRNA-activated kinase or PKR, which is a mediator of the antiviral state and also an inducer of IFN__ (11). It is important to note that the aforementioned activities and functions of the different viral internal products have been characterized using human influenza viruses or avian influenza viruses with the potential to infect humans. It still remains to be determined whether the viral products of other avian influenza viruses perform similar functions or share similar interactions. This proposal will address these important gaps of knowledge.

Agricultural practices contributing to infection with influenza. "Biosecurity is the first line of defense against all avian influenza viruses. Preventing the introduction of avian influenza by eliminating all contact between commercial poultry and wild birds, swine farms, and LBMs is a common, routine and successful practice" (USAHA 2002 Resolution Nº 28.) Unfortunately, implementation of strict biosecurity measures in the LBM systems in the US are difficult because of environmental risks factors and cultural practices that contribute to the perpetuation of viruses in this marketing systems. There are many LBMs in the US, largely concentrated in densely populated areas such as New York (NY), Boston (MA), Chicago (IL), Los Angeles (CA), San Francisco (CA), San Antonio (TX), just to name a few. These LBMs could have a major impact on public health, because HPAI can be a zoonotic disease (18, 85). Intensive education programs are necessary to make owners, suppliers, wholesalers, dealers of commercial poultry and the general public aware of the threat of avian influenza not only for poultry but also potentially to humans. The recent outbreaks of H7 AI in Virginia in 2002 was caused by the introduction of a virus first identified via surveillance in the NY/NJ area and traced to an auction in PA (USAHA and Virginia Dept. of Ag and Cons. Serv.) Likewise, the recent outbreak of H7 viruses in Delaware and Maryland in 2004 can be traced back to viruses circulating in the LBMs. H7 AI viruses were first isolated through sampling conducted in the LBM system of the East Coast in 1994 and efforts to eradicate them have been unsuccessful. A LPAI virus characterized the outbreak in Virginia; where more than 4,000,000 birds were sacrificed to prevent the spread and possible emergence of a HPAI virus. The outbreak caused the state of Virginia huge economic losses. Not controlling the spread of low-pathogenic avian influenza viruses can have devastating consequences as exemplified by the outbreak of H5 in Pennsylvania in 1983-1984 that resulted in the emergence of a HPAI (1, 14, 48, 117). This outbreak led to the declaration of a state of emergency and the implementation of a vigorous eradication campaign. The campaign cost over $63 million in Federal funds and an additional $350 million in increased consumer costs (24). Over 17 million birds died or were slaughtered. The epidemic disrupted the market supply of eggs and poultry meat and impacted also in the cost of pork and beef, which had small increases. Nowadays, a new outbreak of a HPAI such as the 1983 outbreak in Pennsylvania would have an economic impact of up to $120 million to the poultry industry. The costs of not controlling an outbreak would be even more astonishing to the consumers, in excess of $10 billion. It is clearly of interest to devise mechanisms to best contain future epidemics of highly pathogenic AI.

Most efforts at controlling avian influenza in poultry have been focused on the H5 and H7 subtypes; however, control measures should arguably be in place to diagnose and control influenza viruses regardless of the HA subtype. LPAI H9N2 and H6N1 strains circulating in LBMs served as gene donors in the reassortments leading to the emergence of the 1997 Hong Kong H5N1 in China (35, 42, 86). This virus infected at least 18 people; killing six (19, 20, 65, 99). The continuous circulation of non-pathogenic non-H5, non-H7 LPAI in poultry markets of China has contributed to the subsequent outbreaks of highly pathogenic H5N1 viruses in 2001, 2002, 2003, amd 2004, including two more human infections, one of them fatal. Fortunately, these H5 viruses have not spread from person-to-person, although major surveillance efforts in LBMs in China continue to monitor closely the emergence of such a virus. Poultry markets around the world must be considered convenient sampling sites to detect the presence of influenza viruses that can be devastating not only for poultry but also to humans. In the US, Suarez et al. have shown that the LPAI H7 viruses circulating in the LBM system of the Northeastern States have undergone repeated reassortment with viruses of other subtypes (95). It is possible that these reassortments may have contributed to successfully maintaining H7 viruses in the markets. Interestingly, LPAI H7 viruses have circulated in the NY/NJ markets for many years without - to our knowledge- acquiring highly virulent properties. These considerations underscore the importance of interspecies transmission of waterfowl AI to poultry, and alert to the large gap of knowledge regarding selection forces and molecular mechanisms that lead to the emergence of highly pathogenic viruses.

Diagnostic tools and vaccines against avian influenza. The diagnosis of avian influenza is straightforward: virus isolation in eggs and agar gel immunodiffusion assay, although lengthy, are considered the gold standards (21, 25, 83, 93, 108). For surveillance and prevention and control of outbreaks; however, more rapid and highly sensitive methods are needed. Ideally, the methods should be easy to perform and amenable for use on site. Nucleic acid-based detection (PCR) and protein-based detection (antibody or antigen) are available for early diagnosis of avian influenza, although sensitivity, reproducibility, and reliability are major issues for these methods to replace the gold standards. Real time reverse transcription PCR (RRT-PCR) has been used for the detection of influenza and it was adopted by NVSL as a key diagnostic test during the H7 outbreak in Virginia in 2002 and it was readily used during the 2004 outbreak in Delmarva (93). A great advantage of this system is that it can be modified to detect influenza viruses of all known subtypes and/or specific subtypes depending on the need. However, widely implementing this method in the LBM system and/or in small producer sites is difficult due to the cost of the equipment, reagents and the need for technically trained personnel.

There are five commercial-available rapid influenza antibody-based detection kits: Directigen Flu A (Becton Dickinson), QuickVue Influenza test (Quidel), FLU OIA (Thermo Biostar), Zstat Flu (ZymeTx, Inc.), and Now Flu A (Binax, Inc.). These immunoassays are designed to detect influenza A virus and to provide a result within 10 - 30 minutes of set up; however they have been optimized to detect influenza (nucleoprotein or neuraminidase) from human samples (nasal swabs or gargles) and cannot discriminate among subtypes. Recently, Lu et al developed a dot-ELISA assay specific for avian influenza. The sensitivity and specificity of this method is comparable to the Directigen kit with the added advantage of being specific for H7 viruses; however it is less sensitive than virus isolation from eggs (59, 109). To minimize the use of virus isolation in eggs and replace it with any other, faster method for diagnosis, the latter has to be equally sensitive, reproducible and reliable in different laboratories and/or market and farm settings. Taking advantage of the integrated nature of this proposal, we intend to develop, validate, and implement, in diagnostic laboratories across the country, more accurate, sensitive, reproducible and simpler tests specifically designed for early detection of avian influenza. We will concentrate on techniques to identify the H5, H6, H7 and H9 subtypes, which have been associated with major economic losses in poultry in the US.

Vaccination is the second line of defense against avian influenza. However, as it has been stated recently (37) "There continues to be an argument about the use of (inactivated) vaccine for influenza control... The important framework for this argument is whether it is better to use an (inactivated) vaccine or to allow mildly pathogenic avian influenza to circulate unchecked." The USDA has licensed few inactivated virion avian influenza vaccines and a recombinant fowlpox-avian influenza H5 HA vaccine (36, 100-102, 119). Field use of non-H5 non-H7 avian influenza vaccines requires approval by the individual state governments; while H5 and H7 vaccines can only be used as part of an official USDA program and require approval by the federal government (36). Alternative approaches for avian influenza vaccines have included the use of subunit HA protein, DNA immunization and live recombinant vaccines (12, 13, 27, 51, 52). These different approaches have shown promising results protecting against disease and death in challenge studies; however, further studies are needed to evaluate vaccine effectiveness in viral shedding and transmission and, most importantly the generation of mucosal immunity. Influenza virus reverse genetics has provided recently a new alternative for making live attenuated or inactivated vaccines (70, 97, 98, 104, 111). Taking advantage of the collaborative nature of this project, we propose to develop and test different vaccination strategies, compare them side-by-side and run pilot tests in an actual LBM setting to assess their efficacy, potency and evaluate their performance to decrease or eliminate viral load.

PRELIMINARY RESULTS TOP
To comply with the page limitations of the proposal, we have chosen to discuss the preliminary results available from the participating laboratories within the introduction to the relevant Specific Aims. Please note illustrations in the appendix that show the relationship among different projects, collaborations and role of the participating investigators.

> Next
RESEARCH DESIGN AND METHODS